This thesis explored service-users' experience of developing self-compassion using Compassionate Mind Training (CMT). A systematic review of the existing literature on compassion and compassion-based interventions identified the primarily focused of research has been on symptom outcome measures utilizing quantitative approaches. No study that explored the experiences and processes of developing self-compassion was identified. The aims and epistemological stance of the researcher led to a qualitative methodology being used. Once ethical approval was gained, six participants were recruited from two CMT programmes, being run by the same facilitators. All service-users who completed the programme reported a development of self-compassion. Participants were interviewed face-to-face using a semi-structured interview schedule. Interpretative Phenomenological Analysis was selected to analyse the transcribed data. The analysis of the data identified four super-ordinate themes: essential components of self-compassion; process of self-compassion development over time; group experiences; and negotiating change. The first two themes focused on the development and experience of self-compassion. The second two themes included the experience of developing self-compassion in a group and the general change process. Each theme is discussed and presented with supporting quotes. The study found that service-users are able to develop components of self-compassion using CMT. However, continued practice of compassionate behaviours needs to follow the end of the programme for self-compassion to become habitual. The study also identified a number of group factors and processes that prevented or promoted therapeutic change. The results are discussed with reference to the current literature on self-compassion, compassion-based interventions, group processes, mechanisms of change, and further investigation. A model of self-compassion development is proposed. Finally, the researcher's critical reflections of the research process are presented.

Background: Iraqi people have been experiencing traumatic events continually for several decades. Consequently, high prevalence rates of trauma-related symptoms have been documented. In contrast, there is a clear lack in mental health services available for traumatised people. This study aimed to screen for PTSD, depression, and anxiety, assess related variables (e.g. coping strategies, posttraumatic cognitions, and social support), and develop a self-help guide (SHG) for traumatised university students. Methods: Self-report scales were validated via two studies. The first study validated three scales, including Baghdad Trauma History Screen (BTHS), Brief Cope, and Social Support in a sample of 360 (140 males, 220 females) university students. In the second study, the psychometric properties of the scales of Posttraumatic Stress Symptoms (SPTSS), Posttraumatic Cognitions Inventory (PCTI), and the Hospital Anxiety and Depression Scale (HADS) were gained in a sample of 505 (199 males, 306 females) university students. The SHG was developed. Its effectiveness was examined in a sample of participants who reported symptoms that fully met PTSD criteria. The sample included 125 participants: 65 in the experimental group (used the guide for six weeks) and 60 in the control group (did not use the guide). Baseline tests were conducted in both groups before conducting the experiment and post tests afterward. In addition, focus groups were conducted with experts and university students to evaluate the SHG. Results: The scales were reliable and valid. Eighty four percent (424/505) of participants reported at least one traumatic event. Thirty five percent of 424 traumatised students fully met the DSM-IV criteria for PTSD with high levels of Depression and anxiety. Only 11% of participants reported no symptoms at all. Females reported more symptoms than males. The results also revealed that active coping, seeking support, and perceived social support from family was associated with low levels of PTSD, while high levels of PTSD related to the number of traumatic events either experienced by participants themselves or their family members or friends, non problem focused coping, negative posttraumatic cognitions about self and the total posttraumatic negative cognitions. After using the SHG, there were significant increases in the tendency to use active and seeking support coping and significant decreases in non problem-focused coping, negative cognitions about self and the total negative cognitions. There was a significant reduction in levels of PTSD. This reduction related to re-experiencing and avoidance symptoms but not hyperarousal. The level of depression was also reduced. Anxiety was not reduced. However, the experiment was not double blind due to practical reasons and this may influence the results. Conclusion: using the SHG enhanced coping strategies and reduced negative cognitions about self. This led to a reduction in PTSD and depression. Several implications were presented. In addition, self-report scales that validated in this study can be used in future research in the field of traumatic stress in Arabic-speaking populations.

Background: There has been relatively little research on the prevalence and use of tobacco products in developing countries, where the majority of morbidity and mortality from tobacco use in this century is expected to occur. This is particularly true of countries in Africa. I conducted this study in the Ashanti region of Ghana, primarily to measure prevalence and risk factors for smoking, and secondarily to develop a template for national surveys in similar settings in developing countries. I also investigated the history of tobacco use in Ghana and looked into current implementation of tobacco control policy, in particular the Framework Convention on Tobacco Control (FCTC). The FCTC Is the World Health Organization's first public health treaty, established to counter the tobacco pandemic internationally. Methods: First, using electronic literature searches of the tobacco document archives and local library, I searched for all documents with information on the tobacco industry in Ghana and all studies of the prevalence of smoking in Ghana. Secondly, using a two-stage cluster randomized sampling design, I collected data from adults aged 14 and over in a representative household sample of approximately 720 households in the Ashanti Region of Ghana. Finally, I conducted interviews with 20 key policy makers involved with Ghana's implementation of the Framework Convention on Tobacco Control (FCTC) and other tobacco control policies to assess Ghana's progress of implementation. Results: Searches of the literature and tobacco document archives established that British American Tobacco (BAT), and latterly the International Tobacco Company Ghana (ITG) and its successor, the Meridian Tobacco Company (MTC), have been manufacturing cigarettes In Ghana since 1954. After an initial sales booming the two decades after independence in 1957, further increases in consumption typical of the tobacco epidemic in most countries did not occur. Possible key reasons include the taking of tobacco companies into state ownership, and a lack of foreign exchange to fund tobacco leaf importation in the 1970s, both of which may have inhibited growth at a key stage of development; and the introduction of an advertising ban in 1982. BAT ceased manufacturing cigarettes in Ghana in 2006. My survey involved 7096 eligible individuals resident in the sampled households, of whom 6258 (88%; median age 31 (range 14-105) years; 64% female) participated. The prevalence of self-reported current smoking (weighted for gender differences in response) was 3.8% (males 8.9%, females 0.3%), and of ever smoking 9.7% (males 22.0%, females 1.2%). Smoking prevalence was strongly related to increasing age, being highest in the 60-69 age-group (Odds Ratio relative to 14-19 year olds 6.36 (95% Confidence Interval 3.26 to 12.38, Ptrend<0.001), and varied significantly in relation to religion (overall p<0.001), being particularly high in those of Traditionalist belief relative to the Christian majority (adjusted OR 7.50, 95% CI 4.43-12.69);in relation to education level (overall p=0.03, adjusted OR for those with no or only primary education compared with those of tertiary education OR 1.49, 95% CI 0.81-2.73); and in relation to occupation (overall p=0.003, adjusted OR for skilled workers relative to the unemployed 0.66, 95% CI 0.41-1.06). Smokers were more likely to drink alcohol (adjusted OR 7.70, 95% CI 4.63-12.93, p<0.001) and to have friends who smoke (adjusted OR 4.24, 95% CI 3.52-5.11 p<0.001), and significantly less likely to take exercise (adjusted OR 0.82, 95% CI 0.72-0.93, p<0.05). Among smokers, over three quarters (76%) had attempted to quit in the last six months, with the main sources of advice being friends and spouses. Use of smoking cessation medications, such as nicotine replacement therapy, was very rare. About 10% of cigarettes smoked were smuggled brands. About a third (38%) of smokers were highly or very highly dependent. Overall the proportion of ever-smokers who had quit smoking was high (61%) in all age groups. The median number and Interquartile range of cigarettes smoked per day by male and female current smokers on weekdays were respectively 6(1- 40) and 5 (4-10), and at weekends 19 (2-70) and 11 (8-20) respectively. The commonest brands smoked were London Brown (42%) and King Size (22%), both manufactured by BAT. Smokeless tobacco had been used ever by 3.2% of men and had been used more by older than younger people (adjusted OR for over 50's relative to 14-19 year olds 2.09 (95% 1.38-3.18, p<0.05, Ptrend =0.006). Knowledge of the health risks of smoking, including passive smoking and its impact on children and non-smokers, was high; radio (74%) and television (28%) were the main sources of such information and advice. Levels of health awareness were typically but not invariably higher in older people, in men, among the more highly educated and in those living in rural areas. There were few restrictions on smoking in public, and most people (38%) therefore worked and/or spent time in places where smoking was permitted. There was very strong support (97%) for comprehensive smoke-free legislation, mainly among Christians and Muslims. Despite the advertising ban, around a third of respondents (35%), particularly in urban areas, had noticed advertising of tobacco or tobacco products. Again radio was the main source of exposure (72%) but some had also noticed advertising on television (28%). The interviews with policy makers showed that they had good knowledge of the content of the FCTC, and reported that although Ghana had no explicit written policy strategy on tobacco control, the Ministry of Health had issued several tobacco control directives both before and after ratification of the FCTC. A national tobacco control bill had been drafted but had not yet been implemented, something which the policy makers needed to happen urgently. Challenges identified included the absence of a legal framework for implementing the FCTC, and a lack of adequate resources and prioritization of tobacco control efforts. Conclusions: Despite rapid economic growth and a sustained tobacco industry presence, smoking prevalence In Ghana was low, particularly among younger people. This suggests that In contrast to many other developing countries, progression of an epidemic increase in smoking has been avoided. Awareness of health risks and support for smoke-free policies were high in Ghana. Exposure to tobacco advertising or promotion was limited, and most smokers reported having tried to quit. Whether these findings are cause or effect of the current low smoking prevalence is uncertain. The likely reasons that I have identified for the low smoking prevalence include an early advertising ban, substantial state intervention in the tobacco industry at a crucial point of growth, socio-cultural factors (particularly religion), the harsh economic environment at a time when the industry was experiencing growth and other public health interventions such as health education by stakeholders involved in tobacco control. Although policy makers were aware of the FCTC, implementation of the World Health Organization (WHO) treaty has been slow, requiring an urgent need for the passage of the national tobacco control bill into law to enable the country to sustain its tobacco control efforts.

Background In England there is a comprehensive framework of tobacco control policies to reduce smoking-related harm. Policy evaluation helps to ascertain how policies may be improved so that they have the greatest impact; ineffective policies can be dropped or improved, while effective policies can be kept and improved further in order to optimise their impact. The evaluation of tobacco control policy requires high quality and timely data on smoking and smoking cessation behaviour. Time series analysis (TSA) is a robust way of evaluating policy, as it takes existing trends into account, but requires frequently collected data in large samples over long time periods. The aims of this thesis were to investigate the suitability of a range of existing data sources for evaluating the impact of tobacco control policies in England on quitting behaviour, validate potentially suitable measures, and use validated measures to evaluate the impact of recent tobacco control initiatives in England using TSA. Methods A range of data sources which provide information on smoking cessation behaviour were analysed to determine their adequacy for evaluating tobacco control policies, and previously unvalidated measures were validated. Different approaches to TSA – interrupted time series analysis and multiple time series analysis - were employed to evaluate the impact of the introduction of a new smoking cessation medication, varenicline, the broadening of the indications for nicotine replacement therapy (NRT) to include people with cardiovascular disease and adolescents, and the effect of anti-tobacco mass media campaigns on quitting behaviour. Results Two key indicators of quitting behaviour are quit attempts and smoking prevalence; however, there are currently no frequently collected data from large enough samples covering a long time period on these measures. Survey data are generally not suitable for policy evaluation because they are infrequently carried out and often have small sample sizes, making it difficult to detect small and transient changes in behaviour. In contrast, routine sources of data such as electronic medical records data are often frequently collected in large samples over long time periods. A validation study showed that primary care data from The Health Improvement Network are an accurate source of data on prescribing of smoking cessation medication. Time series analyses of these data showed that both the introduction of varenicline, and the broadening of the indications for NRT, did not increase rates of prescribing for smoking cessation medication. Another study found that tobacco control mass media campaigns appear to be more effective at triggering quitting behaviour than pharmaceutical company NRT campaigns. Conclusions Although there are significant gaps in the existing data available, there are some high quality time series data which can be used to evaluate the impact of tobacco control policies in England. There is a need for regular collection of data on key indicators of quitting behaviour, and the use of time series analysis in policy evaluation can play a vital role in strengthening the evidence for the effectiveness of policies, both in tobacco control, and in other areas of public health. The time series studies in this thesis suggest that recent changes to the availability of pharmacological smoking cessation aids have not had a significant impact on public health, and that recent cuts in tobacco control advertising are likely to have reduced quitting behaviour. .

Tobacco smoking is the leading avoidable cause of death and disability in the world. The UK is unique in that it offers a dedicated smoking cessation service providing behavioural support to all smokers, freely available through the National Health Service and with pharmacotherapy available at prescription cost. The service has been proven effective and cost-effective, yet only a small minority of smokers are currently using these services. The research in this thesis examines smokers’ use of support and how more smokers might be identified and encouraged to use it. The first study investigated whether proactive identification of smokers in a primary care setting and referral into such services is a potential means by which awareness and use of services may be increased. As such a study is reliant on the identification of smokers from primary care records and the accuracy of this data, a precursor to this study investigated the completeness and accuracy of smoking status recording in primary care medical records. General practices in this study had a smoking status recorded for between 42.4 and 100% of patients, and comparison of medical records with responses to self-completion questionnaires revealed that this recording is likely to be inaccurate in approximately 20% of cases. Even so, approximately 40% of smokers who responded to the questionnaire were interested in speaking to a smoking cessation advisor when asked, indicating that there is potential to intervene with smokers identified in this way in primary care and that there is a need which is currently not being met. In the trial, all smokers in 12 intervention practices were proactively identified and offered referral into evidence based support, and compared to 12 ‘usual care’ control practices, a significantly greater proportion of these smokers reported using local smoking cessation services (16.6% and 8.9% respectively). Validated 7-day point prevalence from smoking at 6 months was higher in the intervention than the control groups, although this difference was not statistically significant (3.5% and 2.5% respectively). Post-hoc analysis in the sub-group of smokers who had initially reported that they wanted to speak to a smoking cessation advisor did, however, reveal a significant difference between intervention and control groups (4.0% and 2.2% respectively). A proactive approach to enrolling smokers in smoking cessation services is, therefore, effective if you can identify smokers who want support for their quit attempt. Use of an NHS support service traditionally involves some degree of pre-planning. Anecdotal evidence from the proactive trial indicated that many smokers did not pre-plan their quit attempts and as recent evidence from elsewhere has indicated that a large proportion of smokers make an attempt to quit smoking without any pre-planning, this may in part explain the relatively low proportion of smokers accessin0g services. The next study therefore was a questionnaire survey designed to investigate the occurrence, determinants and use of support in planned and unplanned quit attempts. The study findings revealed that over one third of quit attempts were made without pre-planning, and over half of these unplanned attempts were made without the use of any support and unplanned quit attempts appeared to be more likely to be successful, in line with previous findings. However, the use of evidence-based support is known to increase the likelihood of a quit attempt being successful and thus for each successful unplanned and unsupported quitter there are likely to be many more who are unsuccessful. There has been no detailed exploration of how unplanned quitters engage in quit attempts, why they may or may not choose to use support and their attitudes to the support currently available. Gaining a greater insight into these factors may result in the identification of better ways to support those who make unplanned quit attempts. The final study therefore involved qualitative research with a group of unplanned quit attempters and revealed that smokers’ reports of ‘unplanned’ quit attempts may indeed involve elements of planning and delay, and often this delay is in order to gain access to cessation support. The majority of smokers and ex-smokers interviewed were receptive to the idea of support being immediately available whether or not their last quit attempt had involved support. Engaging smokers in using support at an appropriate time, without the need to delay their quit attempt in order to achieve this, may be a potential means of increasing smokers’ uptake of effective cessation support and subsequently improving quit rates. It is therefore important to investigate ways in which smoking cessation support can be made available to potential quitters within a much shorter timescale.

Background: Integrated Care Pathways (ICPs) are prearranged processes of care which are being increasingly used to deliver mental health services. The literature reveals difficulties in their development and implementation, and a lack of empirical evidence to support their use. Aims: The aim of this research was to investigate how an ICP has been used to manage mental health care in one selected mental health Trust in England. Methods: A case study approach was adopted with several units of analysis. The views of healthcare professionals using semi structured interviews; the experiences of service users and carers using focus groups; contrasting hospital episode and performance statistics with a comparison Trust and documentary analysis of the ICP. Findings & Discussion: Of the healthcare professions, only nurses used the ICP. No professionals used the ICP to support clinical decision making and risk management. However, just over two-thirds (67.2%) of the interventions described in the pathway were delivered. There was no statistically significant difference when comparing performance indicators for an equivalent episode of care between the ICP Trust and non ICP Trust. Service user and carers' experiences revealed that peopled did not feel that their care was individualised to them, although amongst them they had different perceptions of the care process. Conclusions: Mental health ICPs need to reflect the relationships between stakeholders, variability of illness and individual ways of living if they are to provide a framework for managing care in the future that accords with the needs of people using mental health services.

The central issue addressed in this thesis was the validation of a novel risk assessment system designed to meet the security requirements of a high secure forensic hospital. To this end, three research strands investigating the clinical, predictive and preventative utility of the system were pursued. This thesis reports the process of the development, adaptation and implementation of a system based on the HCR-20, called the Structured Clinical Judgement: Risk (SCJ: Risk). The first research strand involved investigation of the processes necessary implement a system of structured professional judgement. The conceptual and operational utility of the system were investigated, and successful implementation of the system within clinical practice was demonstrated by compliance of use by clinical teams. A survey was conducted, investigating the perceptions of clinicians in relation to the clinical utility and usability of the SCJ: Risk. Overall acceptance of the pilot and implementation phase of the system was demonstrated, and the system was perceived to assist clinical teams to structure and document risk-related decisions. The second research strand studied the validity of the SCJ: Risk in predicting intra-institutional behaviour. A prospective investigation of the application of the SCJ: Risk to a forensic population detained in conditions of high security was conducted and this demonstrated variation in the predictive utility of the system. The predictive accuracy of subscale items of the SCJ: Risk, and the individual risk factors comprising the subscales of the system were robust for behaviours relevant to violence and suicide/self-harm. However, prediction of behaviours relevant to escape/abscond, vulnerability to risk from others and subversion of security were not demonstrated. The third research strand involved exploration of the preventative utility of the system, specifically if the identification of a patient as high-risk would minimise the occurrence, or prolong the time to an incident of intra-institutional behaviour. Results illustrated the efficacy of the SCJ: Risk system in the identification of high risk individuals for behaviours relevant to any intra-institutional infraction, violence and self-harm. Patients identified as high-risk displayed a higher prevalence and earlier incidence of relevant intra-institutional behaviour. It is concluded that the system of Structured Clinical Judgement: Risk (as an adaptation of the HCR-20) contributes positively to the overall goals of clarity of risk communication, identification and management of high risk patients within forensic psychiatric practice. However, the extended risk assessment of the SCJ: Risk does not make a significant additional contribution to the parental system on which it is based.

Background: The common argument used against the implementation of tobacco control policies is that revenue from tobacco duty is considerably higher than the health care costs smoking imposes on society. This point is true as revenue in the United Kingdom (UK) totalled £9.1 billion while recent costs estimates for the treatment of smoking-attributable disease totalled £5.2 billion to the UK National Health Service. However, this argument becomes unclear when indirect costs such as productivity loss or cost of absenteeism are incorporated. In the UK, there were 29.2 million employed adults in 2011 of which 20% were current smokers. This equates to approximately 5.84 million employed adult smokers. There are currently no studies which have quantified the economic impact of smoking-attributable indirect costs to both employers and the wider society in the UK. These costs are suspected to impose a large economic burden to society but the best practice methodology for estimating indirect costs and the magnitude of these costs are still unknown. Therefore, the aims of this thesis were to quantify the economic impact of smoking-attributable indirect costs due to productivity loss from premature mortality and absenteeism of workforce and to evaluate workplace interventions which could potentially decrease the burden of smoking in the workforce in the UK. Methods: A number of methods were used along with a range of data sources which provided the information to quantify the economic impact of smoking in the workforce. Cost-of-illness methodology based on the human capital method was utilised to quantify the monetary burden of smoking in the workforce due to premature mortality in the UK. Systematic review and meta-analysis was used to examine the epidemiological association between smoking and absenteeism while also providing the necessary parameters to estimate costs of absence in the UK. Finally, decision analysis and Markov simulation modelling was used to conduct both cost-benefit analysis and cost-effectiveness analysis from the employer's perspective for evaluating workplace smoking cessation interventions of brief advice, individual counselling and nicotine replacement therapy with individual counselling. Results: Cost-of-smoking modelling estimated that smoking was responsible for 96,105 deaths (58% male) in adults aged 35 years and over (17% of all deaths) in the UK annually, resulting in 1.2 million years of total life lost and 357,831 years of productive life lost valued at £4.93 billion in 2010. From the systematic review of 29 longitudinal studies, current smokers had a 33% increase in risk of absenteeism and were absent for an average of 2.74 more days per year compared with non-smokers. Compared with never smokers, ex-smokers had a 14% increase in risk of absenteeism; however, no increase in duration of absence could be detected. Indirect comparison meta-analysis showed that current smokers also had a 19% increase in risk of absenteeism compared with ex-smokers. Consequently, smoking was estimated to cost UK employers £1.46 billion in 2011 from absenteeism in the workplace. Workplace interventions for smoking cessation provide a possible method for reducing the burden of smoking in the workforce. Cost-benefit analysis of workplace interventions resulted in brief advice being the optimal decision strategy for women while brief advice and individual counselling both were optimal decision strategies for men in terms of minimising total costs and maximising return on investment for the employer. If the employer valued maximising quitting instead, cost-effectiveness analysis showed that nicotine replacement therapy with individual counselling would be the optimal strategy given a maximised budget constraint. Conclusion: This thesis has provided the first indirect cost-of-smoking study quantifying the productivity loss due to premature mortality and absenteeism in UK; the first systematic review and meta-analysis which has explored the association between smoking and absence from work; and the first cost-benefit and cost-effectiveness analyses of workplace interventions for smoking cessation in the UK. The implications of this research have particular relevance for UK policy makers and employers to justify stronger tobacco control policy which promotes smoking cessation. However, these findings are not unique to the UK. The thesis has provided the framework and methodology for studies that can strengthen the evidence-base around the economics of smoking in other countries as well.

The current study sought to explore how, if at all, people construct an understanding of the origin and maintenance of their experience of hearing voices. An exploratory qualitative method, social constructionist grounded theory, was adopted throughout the research process. Theoretical sampling was utilised and eight participants who hear voices, or previously heard voices, and were distressed by this experience, were recruited through adult mental health teams. Each participant engaged in one digitally recorded interview. A cyclical process of data collection and analysis was undertaken, utilising Charmaz’s (2006) guidelines throughout. Three overarching descriptive categories were constructed regarding participants understanding of the development and maintenance of hearing voices which included; ‘Search for meaning’, ‘View of self’ and ‘Explanations for voices. A sentence summary of the ‘essence’ of the developing grounded theory constructed is stated below: Participants attempted to construct an understanding of their voices through drawing on three main frameworks (inter, intra and para-personal), but the relative ‘success’ of this pursuit, and potential usefulness of an understanding, is effected by the sense of agency, stigma and hope(lessness) perceived by the individual. This study highlighted participants’ attempts to search for meaning of their voices, but the utility of this was often linked to the hopelessness they experienced, and relatively few participants held an explicit theory of the development and maintenance of their voices. This research offers a unique and distinct contribution to the current literature through illustrating how voice hearers actively searched for meaning in relation to their voice-hearing experience. This highlights the importance of helping people engage in meaning-making processes to help individuals understand the experience of hearing voices. Furthermore, the implications of imposing one theoretical framework, which may be incongruent to the voice-hearers own understanding, to the experience of hearing voices are discussed. The study identified a number of clinical implications, such as the role of psychological formulation in generating a shared understanding of the voices. A number of methodological difficulties were encountered during the research process and are discussed. Future research is warranted to explore voice-hearers from a wider range of cultural, religious and spiritual backgrounds and to explore whether the experience of developing a shared framework to understand their voice hearing is valued.

In the field of personality psychology there is some consensus among researchers that human personality, at the broadest level, can be described in terms of five fundamental personality dimensions. Universally, these personality dimensions are referred to as the “Big Five” model or the Five Factor Model (FFM) of personality: Extraversion, Conscientiousness, Agreeableness, Emotional Stability versus Neuroticism, and Intellect or Openness to Experience. However, currently, there is little conceptual or empirical consensus about a meaningful taxonomy of lower-order facets that make up each of the Big Five personality factors. This thesis sought to identify a parsimonious and replicable taxonomy of lower-order facets of the Big Five personality factors, and test the lower-order facet structure for construct and criterion-related validity. Based on the US Eugene-Springfield community sample (ESCS) (N =375), Study 1 examined facet scale scores from nine widely used personality inventories using Exploratory Factor Analysis in order to identify a shared overall lower-order structure for each of the Big Five personality domains. Factor analyses of 162 facet scales revealed 29 facets for the Big Five which demonstrated good convergent validity. However, some facets (e.g. traditionalism, peacefulness, trust) showed less clear patterns of discriminant validity, and thus appear to be compound traits or blends of two or more Big Five factors. In Study 2, a new 232-item Big Five instrument, the Hierarchical Personality Assessment Questionnaire (HPAQ), was developed to measure the 29 lower-order facets derived in Study 1. In the development phase of HPAQ, the factor scores for the 29 facets from Study 1 were correlated with the International Item Pool (Goldberg, 1999) in order to generate an initial pool of items. An initial pool of 348 IPIP items to mark the 29 facets was then administered to a large sample of undergraduate and postgraduate students at a University in the English-speaking Caribbean (N = 778) with the intention of choosing 8 items that best marked each of the facets. Additionally, the HPAQ was validated in a second sample of undergraduate and postgraduate students (N = 807) against the NEO-PIR and its psychometric properties were further examined. The development and validation of the HPAQ was a first step in moving towards Study 3. Study 3 investigated the differential criterion-related validity of the 29 HPAQ lower-order facets in the prediction of job performance criteria (task performance, counterproductive work behaviour, and organisational citizenship behaviours). In addition, the incremental validities of the 29 lower-order facets in the prediction of job performance criteria were also examined. Overall, Study 3 found that the 29 lower-order facets demonstrated differential criterion-related validity and provided incremental validity beyond the global Big Five factors in predicting the job performance criteria and vice versa. Overall, this thesis empirically derives an initial taxonomy of lower-order facets of the Big Five personality factors based on nine personality inventories and developed a new Big Five personality instrument to measure explicitly this lower-order facet structure. The theoretical and practical implications of these results, limitations, and suggestions for future research are discussed.